Massive protocluster of a periodic maser source G188.95+0.89

ABSTRACT

1 INTRODUCTION

High mass star-forming regions (HMSFRs) is an ongoing topic of open debate as it relates to mass growth (Motte, Bontemps & Louvet 2018). The time-scales of phases associated with this process they state are of the order of 10⁴–10⁵ yr. Observation of such phases is very difficult. A unique property to HMSFRs is the presence of class II methanol (CH₃OH) masers in their early formative phases (Caswell, Vaile & Ellingsen 1995a; Breen et al. 2013). These class II CH₃OH masers have proven to be a reliable signpost of the very early stages of high-mass star formation (Menten 1991; Caswell 1996). Recent results of monitoring related infra-red sources and class II methanol (CH₃OH) masers towards HMSFRs have identified significant accretion events. At present, only four accretion events have been detected, they are: S255IR-NIRS3 (Fujisawa et al. 2015; Caratti o Garatti et al. 2017; Szymczak et al. 2018b), NGC 6334I (Hunter et al. 2017, 2018; MacLeod et al. 2018), G323.46–0.08 (Proven-Adzri et al. 2019; MacLeod et al. 2021b), and G358.93–0.03 (Breen et al. 2019; MacLeod et al. 2019; MacLeod et al. 2019; Sugiyama et al. 2019; Burns et al. 2020).

Another important phenomena associated with HMSFRs are periodic methanol masers discovered by Goedhart, Gaylard & van der Walt (2003). Periodic masers are rare, there are only 25 known so far (Goedhart et al. 2003, 2009, 2014; Goedhart, Gaylard & van der Walt 2004; Araya et al. 2010; Szymczak et al. 2011, 2016, 2018a; Fujisawa et al. 2014; Maswanganye et al. 2015; Sugiyama et al. 2015, 2017; Szymczak, Wolak & Bartkiewicz 2015; Maswanganye et al. 2016; Proven-Adzri et al. 2019; Olech et al. 2020). Only the HMSFR source G323.46–0.08 is periodic and has experienced an accretion event (Proven-Adzri et al. 2019; MacLeod et al. 2021b).

The HMSFR G188.95+0.89 (also known as S 252 or AFGL 5180) is well studied at multiple wavelengths. Oh et al. (2010) reported the parallax distance as 1.76 ± 0.11 kpc, Perseus spiral arm (Reid et al. 2009). The 6.7-GHz CH₃OH masers of this source were discovered by Menten (1991) and reported periodic, τ = 395 ± 8 d, by Goedhart et al. (2004, 2014). Kurtz et al. (2000) report an associated unresolved UCH11 region. Minier et al. (2005) did not detect the continuum radio source, but reported that the masers are projected on a bright mm source with an estimated mass in MM1 of 50 M_⊙. They also detected the presence of CH₃CN and C¹⁸O towards the G188.95+0.89 methanol maser site indicating that they are within hot molecular cores (HMCs) with a gas density |$\geqq$| 10⁵ cm⁻³.

In this paper, we present the results of ongoing 6.7 GHz CH₃OH maser monitoring for G188.95+0.89. Significant variations of maser features are analysed. Also high-resolution mm-wavelength dust continuum and molecular line emission observations are presented.

2 OBSERVATIONS AND DATA REDUCTION

2.1 Single-dish Observations

The radio observations were made using the 26-m telescope of the Hartebeesthoek Radio Astronomy Observatory (HartRAO¹). The 4.5-cm receiver is comprised of dual, cryogenically cooled, RCP and LCP feeds. Each polarization was calibrated independently relative to Hydra A and 3C 123, assuming the flux scale of Ott et al. (1994).

All observations employed frequency-switching (FS) and within 2 h of zenith. Observations completed with a 1-MHz bandwidth and recorded with a 1024-channel per polarization spectrometer (in FS 512 channels are saved). The velocity resolution achieved is 0.044 km s⁻¹ and the typical 3σ root-mean-square (rms) noise per observation was ∼1 Jy. This sensitivity is improved when across several channels, e.g. integration over 10 channels improves the sensitivity to ∼0.3 Jy  km s⁻¹. For all the epochs, half-power beam-width pointing correction observations were carried out. The rest frequency was corrected for the Local Standard of Rest (LSR) velocity v_LSR = +10.0 km s⁻¹. Listed in Table 1 are the parameters of the telescope, receiver, and observation set-up for each epoch.

Observations were made between 2003 June 30 and 2021 January 18 with a cadence of between 10 and 20 d; the cadence varied during several flares. No spectroscopic observations were taken between 2008 September and 2010 December when the 26-m antenna underwent repairs (Gaylard 2010). The position employed at HartRAO is RA = 06^h 08^m 53|${{_{.}^{\rm s}}}$|3 and Dec. = +21° 38′ 30″ (J2000).

2.2 ALMA Observations

We obtained ALMA band 6 archival data on G188.95+0.89 (Project ID:2015.1.01454.S) taken on 2016 April 23 (42 antennas of the 12-m main array), 2016 September 17 (38 antennas of the 12-m main array). Observations of G188.95+0.89 (phase tracking centre at RA_(J2000) = 06^h 08^m 53|${.}^{\rm s}$|3 and Dec._(J2000) = +21° 38′ 30″) were carried out at 1.3 mm (230 GHz) with ALMA band 6 in 2015 using two different configurations. The total time on-source was 19 and 30 min, respectively, and the projected baselines ranged from 15–2600 m. The ALMA correlator was configured to cover nine spectral windows (spws). The raw visibility data were calibrated using the Common Astronomy Software Applications (CASA 5.4) standard calibration and imaging tasks. Bandpass and flux calibration were conducted using the sources J0510−1800 and J0750−1231, respectively. J0603−21591 was used as the gain calibrator. The calibrated visibility data from the two observation blocks were combined using CASA task CONCAT, and CLEAN was used to produce the images of the continuum (rms of 2.2 mJy beam⁻¹). Also images of molecular line emission (typical rms of 2.2 mJy beam⁻¹) with spectral resolution of 1 km s⁻¹ are produced. In this paper, we will focus on the ¹²CO (2-1), CH₃OH (4_{(2, 2)}-3_{(1, 2)}), SiO (5-4), and C¹⁸O (2-1). For continuum subtraction, only line-free parts of the spectra were used. We used CASA to do all the calibration and imaging of the data cubes as well as self calibration to remove residual phase and flux calibration errors. The line data were then imaged with a robust weighting of 0.5.

3 RESULTS

3.1 6.7-GHz CH₃OH masers variability

In Fig. 1, we present selected spectra from the 17 flare cycles (Fl_n for n = 5 to 21) in the 6.7-GHz methanol maser observations associated with G188.95+0.89 studied here (the original four are published in Goedhart et al. 2004). The spectra presented are the maxima of three of these flare cycles (flare number Fl_n for n = 6, 13, and 20). In this image, it can be seen the evolution of the spectral profile over the 18 yr of observations presented. This profile is certainly comprised of many line blended masers, in particular for  + 10 < v_LSR <  + 11.3 km s⁻¹.

Figure 1.

Spectra of the 6.7-GHz methanol masers associated with G188.95+0.89 observed at the maxima of Fl₆, Fl₁₃, and Fl₂₀. Arrows mark the selected velocity features studied here.

Open in new tabDownload slide

A dynamic spectra is the best way to visualize variability of these 6.7-GHz methanol masers associated with G188.95+0.89, see Fig. 2. Features have weakened others strengthened, most appear periodic and variations in velocity, velocity drifts, are apparent. For the purpose of analysis of these observations five features, two at v_LSR = +10.44 and +10.70 km s⁻¹ in the heavily line merged velocity regime, and others at v_LSR = +8.42, +9.65, and  +11.45 km s⁻¹ are selected. Note, the former two features are likely comprised of multiple masers; the others less likely.

Figure 2.

The dynamic spectrum of the 6.7-GHz methanol masers associated with G188.95+0.89. Dashed lines demarcate the spectra of the maximum in Fl₆, Fl₁₃, and Fl₂₀ and plotted in Fig. 1. No observations were taken between 2008 September and 2010 December during repairs.

Open in new tabDownload slide

Time series plots of the integrated flux density, |$F\mathbf {_{Int_{6.7}}}$|⁠, for each selected feature are shown in Fig. 3. A time series plot of the total integrated flux density, |$F\mathbf {_{Int_{6.7}}}$|(Total), is also shown. For each feature, in each epoch of observation, the velocity associated with the maximum flux density in the stated velocity extent is determined. Linear regression analysis is applied to each velocity time series to determine the velocity drift, and the results are included in this table. Two of the features have measurable velocity drifts, with ‘goodness of fit’ values R² > 50 |${{\ \rm per\ cent}}$|⁠. The apparent drift seen in the v_LSR = +10.70 km s⁻¹ feature (see Fig. 1) may be the result of variations of heavily line merged masers contributing to the feature. The features at v_LSR = +9.65 and +11.45 km s⁻¹ are the best fit by a single Gaussian profile each; both possibly devoid of line merged features.

Figure 3.

Time series plots of |$F\mathbf {_{Int_{6.7}}}$| for five maser features identified in Fig. 1 and labelled in each panel. In (d), the total integrated flux density of all masers associated with G188.95+0.85 is plotted. Note the x-axis in (b) through (e) denote the estimated maxima of each flare, Fl_n for n = 5–21 (for τ ∼ 395 d).

Open in new tabDownload slide

All five features are periodic though it is not obvious for the feature v_LSR = +11.45 km s⁻¹ in Fig. 3. The period of each is determined using, first the programme Period04 developed by Lenz & Breger (2004) and secondly the Lomb–Scargle (LS) periodigram (Scargle 1982). The results are listed in Table 2. The average period for each method is included and are within 3σ standard deviation of each other, τ_Period04 = 395 ± 1 d against τ_LS = 397.6 ± 0.7 d.

Figure 4.

(a) Time series plot of |$F\mathbf {_{Int_{6.7}}}$| for the v_LSR = + 11.45 km s⁻¹ feature. The fitted exponentially decaying function (black dashed line) is included. (b) The residual of the integrated flux density, raw data less the fitted exponentially decaying function, is plotted. Note the secondary x-axis in (b) denotes the estimated maxima of each flare, Fl_n for n = 5–21 (for τ ∼395 d).

Open in new tabDownload slide

After equation (1) is subtracted from the original time series the remaining residual, also plotted in Fig. 4, reveals its possible periodic nature. Note the slight increase after MJD 7500; possibly resulting from variations in the V_LSR = + 10.70 km s⁻¹ feature. The total integrated flux density plotted in Fig. 3 is seen first falling prior to MJD 5000 then rising suggesting the features decaying dominate prior to MJD 5000 thereafter brightening of the  +10.70 km s⁻¹ feature dominates.

3.2 G188.95+0.89 continuum emission

In Fig. 5 (top panel), we show the composite image of G188.95+0.89 in WISE bands 1 (3.4μm: red), 2 (4.6 μm: green) and 3 (12 μm: blue), and ALMA 1.3-mm dust continuum emission. The WISE image show a central infrared source (bright in all three bands) corresponding to the 850-μm SCUBA MM1 object, and a green (4.6 μm) dominant object south of the central object corresponding to the 850-μm SCUBA MM2 object (Minier et al. 2005). G188.95+0.89 is resolved into eight 1.3-mm objects (MM1–MM8) with ALMA. MM1–MM4 and MM5–MM8 objects (Fig. 5, bottom panel) are associated with 850-μm SCUBA MM1 and MM2 objects of Minier et al. (2005), respectively.

Figure 5.

(Top) Composite three-colour WISE image of G188.95+0.89. WISE band 1 (3.4 μm), 2, (4.6 μm), and 3 (12 μm) are represented red, green, and blue. (bottom) ALMA band 6 (black contour, levels  = [2, 4, 10, 20, 50] mJy beam⁻¹) and in the zoom-in, band 7 (grey contour, levels  = [2, 4, 20, 60, 80] mJy beam⁻¹) dust continuum emission of G188.95+0.89. The filled ellipses on the bottom left-hand corner beams of the band 6 (larger ellipse) and band 7 (smaller ellipse). The open circle near MM1 peak indicates the position of the periodic 6.7-GHz CH₃OH masers.

Open in new tabDownload slide

ALMA 1.3 mm MM1 and MM2 are central objects (Fig. 5 bottom zoom-in) and MM1 (the brightest dust continuum object) is associated with the periodic 6.7-GHz CH₃OH maser source. Interestingly, a recent ALMA band 7 observations resolved the continuum core in MM1 into a single object but resolved MM2 into 2 continuum cores (grey contours of Fig. 5 bottom zoom-in). Details of the detected continuum sources and their masses are presented in Table 3.

The integrated flux densities for MM5–MM8, that were also used in the core mass estimates, are not primary beam corrected since MM5–MM8 are located close to the edge of the primary beam and will not be discussed in detail due to limited sensitivity.

3.3 G188.95+0.89 millimeter line emission

A number of thermal molecular lines were detected towards the eight millimeter continuum objects, however in this paper, we will focus only the CH₃OH (4_{(2, 2)} − 3_{(1, 2)}), SiO (J = 2 − 1), ¹²CO (J = 2 − 1), and C¹⁸O (J = 2 − 1) as shown in Table 4.

3.3.1 CH₃OH 4_{(2, 2)}-3_{(1, 2)}

Emission from the CH₃OH (4_{(2, 2)} − 3_{(1, 2)}) line was detected towards all eight cores, see Fig. 6. The emission is strongest towards the southern MM5–MM8 and weakest towards MM4. The nature of the emission towards MM5–MM8 is contaminated by the effect of the decreased sensitivity towards the edge of the primary beam. Beyond detection or non-detection of an emission, no further discussions will be made for these objects.

Figure 6.

CH₃OH 4(2,2)-3(1,2) thermal line channel map with dust continuum overlaid (black contour, levels  = [0.002, 0.004, 0.02, 0.06, 0.08] Jy beam⁻¹) dust continuum emission of G188.95+0.89. The colour scale is the intensity in Jy beam⁻¹.

Open in new tabDownload slide

One interesting feature of the CH₃OH (4_{(2, 2)}-3_{(1, 2)}) emission towards MM2 is the presence of a double peak in the emission at the systemic velocity of MM2. There is CH₃OH (4_{(2, 2)} − 3_{(1, 2)}) emission associated with MM1. Minier et al. (2005) indicated the temperature of the dust of the clump hosting the 1.3 mm ALMA MM1–MM4 sources to be ∼150 K.

3.3.2 SiO (J = 2 − 1)

Emission of SiO (J = 2 − 1) is known to trace shocks especially in star-forming regions. The dominant SiO emission is observed in the north-west region of MM2, see Fig. 7. The emission has a bow-shock morphology in the channel corresponding to the systemic velocity of MM2. There are SiO emission (in some cases, very faint emission) associated with each of MM3–MM8, however, no SiO emission is detected towards the MM1 object.

Figure 7.

SiO (J = 2 − 1) thermal line channel map with dust continuum overlaid (black contour, levels  = [0.002, 0.004, 0.02, 0.06, 0.08] Jy beam⁻¹) dust continuum emission of G188.95+0.89. The colour scale is the intensity in Jy beam⁻¹.

Open in new tabDownload slide

3.3.3 ¹²CO (J = 2 − 1)

The ¹²CO (J = 2 − 1) emission in G188.95+0.89 is complicated owing to the effect of the strongly self-absorbed features seen in the line emission. The other plausible reason for the observed complex distribution is that all the millimeter objects may be driving outflows. If the strong emission east of MM2 observed in the −3 km s⁻¹ channel and the emission west of MM2 in the 9 km s⁻¹ channel are both associated with MM2, then they may demarcate an east–west bipolar outflow in MM2, see Fig. 8. There is also some emission north of MM2 and could signify a second outflow emanating from the object; suggestive of multiple YSOs in MM2. The other mm objects are all associated with ¹²CO (J = 2 − 1) emission.

Figure 8.

¹²CO (J = 2 − 1) channel map with dust continuum overlaid (black contour, levels  = [0.002, 0.004, 0.02, 0.06, 0.08] Jy beam⁻¹) dust continuum emission of G188.95 + 0.89. The colour scale is the intensity in Jy beam⁻¹.

Open in new tabDownload slide

3.3.4 C¹⁸O (J = 2 − 1)

We detected high density tracer, C¹⁸O (J = 2 − 1), towards all MM1–MM8, see Fig. 9. MM1 and MM2 are the dominant sources of the C¹⁸O (J = 2 − 1) emission. Interestingly, the brightest C¹⁸O (J = 2 − 1) channel (3 km s⁻¹ channel) shows a distribution of the C¹⁸O (J = 2 − 1) emission to lie in the interface between MM1 and MM2 cores. The possible implication of this to the observed variability in the 6.7-GHz CH₃OH masers will be discussed in Section 4.1

Figure 9.

C¹⁸O (J = 2 − 1) channel map with dust continuum overlaid (black contour, levels  = [0.002, 0.004, 0.02, 0.06, 0.08] Jy beam⁻¹) dust continuum emission of G188.95+0.89. The colour scale is the intensity in Jy beam⁻¹.

Open in new tabDownload slide

4 DISCUSSION

4.1 Implications of the variability in the 6.7-GHz CH₃OH Masers

Originally, Goedhart et al. (2004) reported the period of G188.89+0.89 was 416 d; this was revised in Goedhart et al. (2014) to 395 d. Here, using the longer time series and from two separate methods, the shorter period is confirmed (τ_Period04 = 395 ± 1 d and τ_LS = 397.6 ± 0.7 d). Durjasz et al. (2021) also confirm, though observing fewer flare cycles, the shorter period, τ = 395 ± 8 d.

Class II methanol masers are only found associated with high mass star-forming regions (Minier et al. 2003). Such regions experience accretion events (identified by flaring masers) such as those reported in S255IR-NIRS3 (Fujisawa et al. 2015; Szymczak et al. 2018b), NGC 6334I (MacLeod et al. 2018), G323.46–0.08 (Proven-Adzri et al. 2019; MacLeod et al. 2021b), and G358.93–0.03 (MacLeod et al. 2019; Sugiyama et al. 2019). However, three of the features in Table 2 are decaying, in particular, the v_LSR = + 11.45 km s⁻¹decays exponentially (Fig. 4). Interesting, historical observations suggest the originally detected peak flux density feature, v_LSR ∼ +10.5 km s⁻¹, has varied markedly since detection, less than 500 Jy before 1993 (Menten 1991; Caswell et al. 1995b), greater than 600 Jy before 2009 (Goedhart et al. 2004; Green et al. 2012) to ∼500 Jy on average here. Note at present the peak flux density is found at v_LSR = + 10.70 km s⁻¹ and not +10.44 km s⁻¹. Caswell et al. (1995b) reported that the feature v_LSR = +11.2 km s⁻¹ decreased slightly from 1991 (∼250; Menten 1991) to 1993 (∼230 Jy). In 1999 August, it was only ∼100 Jy (Szymczak, Hrynek & Kus 2000) while reaching a maximum, ∼160 Jy, in 2001 November (Goedhart et al. 2004); thereafter, the exponential decay is reported here. Above equation (1) cannot describe these variations, nor does it fit the maximum reported in Goedhart et al. (2004). It appears a weak flaring event, increasing by a factor of ∼2.6 in flux density, occurred between 1999 August and 2001 November in the periodic feature at v_LSR = +11.45 km s⁻¹. This decaying periodic feature resembles that shown for decaying periodic maser features in G323.46−0.08 (MacLeod et al. 2021b). Further analysis of historical data, and data other transitions (e.g. 12.2 GHz CH₃OH and/or 22.2 GHz H₂O), are required to confirm.

Spot maps of the 6.7-GHz methanol masers are shown in Minier, Booth & Conway (2000) and Hu et al. (2016); all features reside in a 180 × 180 au box assuming a distance of 1.76 kpc (Oh et al. 2010) about the bright reference feature (v_LSR = + 10.33 km s⁻¹). The weakening features reside within ∼50 au radius of the reference feature, v_LSR = +8.42 km s⁻¹ slightly north while +11.45 km s⁻¹ is slightly south of the reference maser. The strengthening feature (+10.70 km s⁻¹) is found ∼90 au to the south-east from the reference maser. The non-varying feature, v_LSR = + 9.65 km s⁻¹, is further south-east (∼140 au). Minier et al. (2000) suggest that the masers surrounding the reference maser could be part of an outflow from the disc.

It is not clear whether velocity drifts, including those seen here, are caused by variability of spectrally blended masers (Szymczak, Wolak & Bartkiewicz 2014) or by motion of the gas (Goddi, Moscadelli & Sanna 2011). The accretion discs surrounding massive star-forming regions may experience infalling gas. Szymczak et al. (2014) report a velocity drifting feature in Cepheus A; they suggest it may be an artefact of variable line-merged features or the result of masers located in regions of infalling gas. MacLeod et al. (2021a) propose the inner radius of the accretion disc surrounding the protostar of G9.62+0.20E is infalling resulting in systematic velocity drifts of its associated methanol masers. However, they also state that it may be caused by masers in a precessing disc. Above in Table 2, it can be seen that two features experience measurable velocity drift. The blue-shifted feature, v_LSR = +9.65 km s⁻¹ is blue-shifting (−2.38 ± 0.03 × 10⁻⁵ km s⁻¹ d⁻¹) while the other, v_LSR = +11.45 km s⁻¹, is red-shifting (+1.88 ± 0.04 × 10⁻⁵ km s⁻¹ d⁻¹). The drifts are of the order of those seen in G9.62+0.20E (MacLeod et al. 2021a), ranging from −6 to +3 × 10⁻⁵ km s⁻¹ d⁻¹. However, an insufficient number of features with measurable velocity drift are found, thus, rendering any comparison to G9.62+0.20E indeterminate. These two features are found well-separated in the spot maps shown in both Minier et al. (2000) and Hu et al. (2016), by ∼200 au. Perhaps the central features are influenced by an outflow (Minier et al. 2000) and this motion may force the features at the edges to experience velocity drift. Even the now brightest feature, at v_LSR = + 10.70 km s⁻¹, is blue-shifting though this may be due only to variations of heavily line-merged features. All of these periodic masers are associated with MM1. More observation and analysis is required to explain these variations.

4.2 Binary system in G188.95+0.89-MM2

There are a number of indicators suggesting the existence of a binary system in MM2. First, the high-resolution ALMA band 7 dust continuum image resolves MM2 into 2 cores separated by 0.1 arcsec. Secondly, the CH₃OH (4_{(2, 2)} − 3_{(1, 2)}) emission is coincident with the systemic velocity and the spectral profile of emission towards MM2 has two pronounced features (see 3 km s⁻¹ channel of Fig. 6). In Fig. 10, the spectra of ¹²CO (solid lines) and CH₃OH (4_{(2, 2)} − 3_{(1, 2)}) (dashed line) taken from MM2 using a 4 arcsec ellipse are presented. It is important to note that the ¹²CO emission is optically thick. In the ¹²CO spectra (at least), self-absorption features are visible, implying the ¹²CO emission is absorbed by cold foreground material in the region. The three vertical dotted lines in Fig. 10 indicate the prominent absorption feature (middle line at 3 km s⁻¹) and two absorption features (−2 and 7 km s⁻¹) one on each side of the prominent feature. We noted that missing flux issue can give rise to such spectral feature but in this case, the observed absorption feature is real. The most prominent absorption feature in the ¹²CO spectra corresponds in V_LSR with the CH₃OH (4_{(2, 2)} − 3_{(1, 2)}) absorption feature separating the two resolved cores in MM2. All these observed features suggest multiple (or at least binary) young stellar objects in MM2.

Figure 10.

¹²CO (solid lines) and CH₃OH (dashed lines) spectra (of G188.95+0.89 extracted with an ellipse of size ∼4 arcsec centred on MM2. The vertical lines indicate the observed absorption features in ¹²CO line. The multiple absorption features in the ¹²CO emission and the double peak feature in the CH₃OH line point to multiplicity (or at least binary) of driving sources in MM2. Band 7 ALMA dust continuum detected 2 cores in MM2.

Open in new tabDownload slide

Our suggestion of the presence of at least a binary, or multiple, YSOs in MM2 is supported by the complexity of the orientation of the outflowing gas traced by ¹²CO, see Fig. 8. An east–west outflow feature with an axis through the peak position of MM2, as well as a north-west and south-east outflow also intersecting with MM2, can be observed. These support the suggestion at least a binary system is present in MM2. High-resolution continuum and line observations will be required test the multiplicity of YSOs in MM2.

5 CONCLUSION

We presented results of 1.5 decades of monitoring observations of 6.7-GHz CH₃OH masers towards G188.95+0.89 and ALMA band 6 observational results and found the following:

• G188.95+0.89 protocluster is resolved into eight 1.3-mm objects (MM1-MM8). MM1 and MM2 are the central objects and MM1 (the brightest dust continuum object) is associated with the periodic 6.7-GHz CH₃OH maser source.

• MM2 hosts more than one YSO, likely a binary system.

• Strong emission of ¹²CO at the east and west of MM2 point to the presence of an east–west bipolar outflow in MM2. Emissions north of MM2 also suggest a second outflow emanating from the object, which could signal multiplicity of YSOs in MM2.

• All five features are periodic, suggesting a common background source of seed photon. The light curve shape of the maser features are similar to those of Mira variable stars and could suggest pulsation of the protostar as the possible driver of the periodicity (Inayoshi et al. 2013).

• Two CH₃OH maser features are reported with measurable velocity drift.

• Outflows are identified in ¹²CO (2-1) line emission. While no direct detection of accretion discs was possible with the current observations, the detected outflows suggest the presence of accretion discs in the source. Accretion disc may experience infalling gas and velocity drifts may be due to an artefact of variable line-merged features or by infalling gas.

ACKNOWLEDGEMENTS

MMM acknowledges the support of The Technical University of Kenya, NRF-South Africa, HartRAO-South Africa, and the Newton Fund. This paper makes use of the following ALMA data: ADS/JAO.ALMA 2015.101454.S. ALMA is a partnership of ESO (representing its member states), NSF (USA), and NINS (Japan), together with NRC (Canada) , MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ. This publication makes use of data products from the Wide-field Infrared Survey Explorer, which is a joint project of the University of California, Los Angeles, and the Jet Propulsion Laboratory/California Institute of Technology, funded by the National Aeronautics and Space Administration.

DATA AVAILABILITY

• This paper makes use of the following ALMA data: ADS/JAO.ALMA 2015.101454.S. and can be accessed on the ALMA Science portal.

• This publication makes use of data products HartRAO 26-m maser monitoring project and will be made available on a reasonable request.

• The WISE data used in this publication can be access on the IRSA public data archive.

Footnotes

REFERENCES